Eliminated last exit() call from libmysqld.

MDEV-14545 Backup fails due to MLOG_INDEX_LOAD record

- Changed the unsupported_redo test case to avoid checkpoint
- Inserting more rows in purge_secondary test case to display evict monitor.

Unlike commit a54abf01 claimed,
the caller of THD::awake() may actually hold the InnoDB lock_sys->mutex.
That commit introduced a deadlock of threads in the replication slave
when running the test rpl.rpl_parallel_optimistic_nobinlog.

lock_trx_handle_wait(): Expect the callers to acquire and release
lock_sys->mutex and trx->mutex.

innobase_kill_query(): Restore the logic for conditionally acquiring
and releasing the mutexes. THD::awake() can be called from inside
InnoDB while holding one or both mutexes, via thd_report_wait_for() and
via wsrep_innobase_kill_one_trx().

            the non-recursive CTE defined with UNION

The problem appears as the columns of the non-recursive CTE weren't renamed.
The renaming procedure was called for recursive CTEs only.

To fix it in the procedure st_select_lex_unit::prepare
With_element::rename_columns_of_derived_unit is called now for both CTEs:
recursive and non-recursive.

ha_innobase::unlock_row(): Use a relaxed version of the
trx_state_eq() debug assertion, because rr_unlock_row()
may be invoked after an error has been already reported
and the transaction has been rolled back.

By definition, c_lock->trx->lock.wait_lock==c_lock cannot hold.
That is, the owner transaction of a lock cannot be waiting for that
particular lock. It must have been waiting for some other lock.
Remove the dead code related to that. Also, test c_lock for NULLness
only once.

The caller of THD::awake() should never hold any InnoDB mutexes,
so we can always acquire lock_sys->mutex and trx->mutex.

MDEV-13549 Fix and re-enable test galera.galera_var_slave_threads

Two changes were made to the test:

1) Suppress warning "Refusing exit for the last slave thread."
   This warning was already suppressed, but on the wrong node.

2) The test occasionally fails because it expects that the
   number of applier threads changes immediately after
   changing the value of ```variable wsrep_slave_threads```.
   Which is not true. This patch turns snippets like this:

   ```
   SET GLOBAL wsrep_slave_threads = x;
   SELECT COUNT(*) = x FROM INFORMATION_SCHEMA.PROCESSLIST WHERE USER = 'system user';
   ```

   Into proper wait_conditions:

   ```
   SET GLOBAL wsrep_slave_threads = x;
   let $wait_condition = SELECT COUNT(*) = x FROM ...;
   --source include/wait_condition.inc
   ```

- Work around for MDEV-13942: Drop spatial index to avoid possible hang

MDEV-15540 fix for MTR tests using wsrep_recover.inc

MDEV-14144 Fix for MTR tests complaining "Query didn't return a result set"

MDEV-13549 Fix for test galera.galera_var_max_ws_rows

MDEV-13549 Fix and re-enable MTR test galera.MW-366

MDEV-13549 Fix and re-enable test galera.galera_bf_abort_for_update

MDEV-13549 Fix and re-enable test galera.MW-286

These tests were also subject to error "Wrong usage of mutex
'LOCK_wsrep_thd' and 'LOCK_thd_kill'".

"mtr func_date_add" failed on 32-bit platforms. Removing a wrong case to "long".
Both values[] and log_10_int[] are arrays of "ulonglong", no cast is needed.

add the test case (the bug was fixed in d390e501)

Refactor get_datetime_value() not to create Item_cache_temporal(),
but do it always in ::fix_fields() or ::fix_length_and_dec().

Creating items at the execution time doesn't work very well with
virtual columns and check constraints that are fixed and executed
in different THDs.

while it should look at the actual field_type() and use get_date()
or get_time() as appropriate.

test case is in the following commit.

Do not assume that it's always item->field_type() - this is not the case
in temporal comparisons (e.g. when comparing DATETIME column with a TIME
literal).

It's a generic function, not using anything from Arg_comparator.
Make it a static function, not a class method, to be able to use
it later without Arg_comparator

will be used in following commits

reorder items in args[] array. Instead of

  when1,then1,when2,then2,...[,case][,else]

sort them as

  [case,]when1,when2,...,then1,then2,...[,else]

in this case all items used for comparison take a continuous part
of the array and can be aggregated directly. and all items that
can be returned take a continuous part of the array and can be
aggregated directly. Old code had to copy them to a temporary
array before aggreation, and then copy back (thd->change_item_tree)
everything that was changed.

musl ships the header for other purposes, but makecontext is not
implemented. fix the check to detect if makecontext is implemented
before enabling code using it.

Test passes reliably after applying missing commit from upstream.

The test was not deterministic as it would not wait for the second
INSERT to be applied (and thus the victim transaction rolled back).

- Fixing the windows failure of unsupported_redo test case.
mariabackup --tables-exclude option only restricts ibd file.